Thread by @swardley

X : I know you says it's a commodity but we've decided to custom build this software. It creates a better brand experience.
Me : Do you know how much that is costing you?
X : It's just engineering time.
Me : Oh god ...

Me : Do you map?
X : No.
Me : Ok, let's pretend we're in the business of being a Tea Shop, selling to the public. Here's a basic map.

... at the top of the map you have the anchor (our focus) which is the business (that needs to sell cups of tea) and the public (that has a need / desire to drink them - we hope). The maps has the basic components involved. The first thing you do is share the map ...

... others will hopefully chip in with all the components you missing i.e. staff. We each usually only have part of the picture, maps are good for helping us communicate and gain a better map.

Of course maps can also help us anticipate future changes e.g. staff becoming robots etc. At least we can discuss it. We might disagree. The map isn't right, it's a way for each of us to challenge each others assumptions.

So currently we have strategy (robots!). business (the value chain), HR (staff) and sales (the needs) talking with each other. Someone from operations is bound to chip in going "why we bloody building custom kettles"

At which point someone from marketing will probably mutter a thing or two and branding and exclusivity being part of the user needs. Well, at least we're now all talking.

Hence with finance we start to add metrics (each of the nodes are stock of capital, the lines are flows), some values to core value chain ... we'll come back to the branding bit later ...

from this we can build a basic income statement around each cup of tea and at least we're all talking ...

Now, we can go back and look at the cost of custom building (i.e. the depreciation on the kettle) vs a more industrialised component. We now know what that branding / exclusivity / whatever is costing us. Basically, it's almost twice the profit, so you'd have to question why?

X : That's all very good but I don't see what cups of tea and maps have to do with custom building software?
Me : Really? Do you know what Lambda is?
X : Yes
Me : You've heard of billing per function?
X : Yes
Me : Well ...

Me : Let's pretend we're an online taxi firm, we've built our user app in Lambda and to make life easy, I'll make it look like the cup of tea ... I can build income statements through the application (this is what capital flow is all about) ...

so now, we can have the same "custom building kettle" conversation around brand exclusivity but this time over a home grown function called "car.getProfiles()" ...

... far too much of IT is custom built stuff that doesn't need to be. Once we get the tools in place and start monitoring capital flow, this will change and these practices will spread to the rest of the business (which is also hopeless with bias, duplication etc) ...

... this is why you're going to get the $1bn company that is just one or two people providing a single function. The industry will have been custom building that function all over the place. Lambda will help clean this up, once the tools / monitoring are there ...

... this doesn't even cover wasted human talent, speed and all the other nasty side effects that go into rebuilding the same thing over and over and over. The overwhelming output of most business is waste. Serverless is way larger than you think. More significant than cloud was.

This is what I was doing with Zimki all those years ago. It leads to a world of worth based development & entire new investment strategies. It changes everything that IT touches, which is basically ... everything.

Buckle your seatbelt Dorothy, 'cause Kansas is going bye-bye.